To: David S.
From: Geoffrey Klempner
Subject: Essay on Hume's Fork
Date: 17 March 2004 13:09

Dear David,

Thank you for your email of 8 March, with your first essay for the Metaphysics program, in response to the question, 'If we take in our hand any volume, of divinity or school metaphysics for instance; let us ask, 'Does it contain any abstract reasoning concerning quantity or number?' No. 'Does it contain any experimental reasoning concerning matter of fact and existence?' No. commit it then to the flames: for it can contain nothing but sophistry and illusion' (David Hume 'Enquiry Concerning Human Understanding', Section XII, Part III). - Comment.

This is a good answer. There are two main ways to go with this question. The first is to consider Hume's fork as interpreted by contemporary critics of metaphysics who do not subscribe to strict Humean empiricism, the second is to consider how Hume himself understood the challenge, and its bearing on his account of the scope and limits of human knowledge.

You have chosen the second route. Hume sets out to challenge metaphysics. However, it turns out that the restrictions which he places on the acquisition of knowledge are so strict that he is unable even to account for knowledge of the external world.

Hume is prepared to allow, as empirical knowledge, anything that can be deduced, using logical principles alone, from what he regards as the basic empirical 'data', namely our immediate, subjective sense impressions. However, as you state, 'a reality that depends only on observations has no mechanism for generating a world that exists independently of those observations.' This is a crushing objection to Hume's project.

The best illustration of this is Hume's discussion of the continued and distinct existence of objects (in the section of the 'Treatise' entitled, 'On Scepticism With Regard to the Senses'). Sense impressions plus logic cannot generate so simple a thing as a spatio-temporal existent. The homely observation that when you press one of your eyeballs you see double, undermine the very idea that what we see with your eyes are objects that would be just as they are if we were not looking at them.

You have chosen the problem of causality. The case is a little less clear here, because Hume as pretty workable account of causality in terms of the logical notion of universal generalization. Contemporary philosophers who reject Humean empiricism are still debating whether Hume was right about causation. If you are 'Humean' about causation, then what, 'The stone broke the greenhouse window' *means* is that for all x, at any place or at any time, if x satisfies precisely the same initial conditions as this particular stone-throwing event, then x is also a window-breaking event. No mysterious 'causal influence'. Instead, we have a claim that can never be conclusively verified. But at least we know what is being claimed.

Kant's argument in the first part of the Critique of Pure Reason attacks both questions, the question of spatio-temporal existents and the question of causality. His response is that the concept of space-occupying, temporally persisting substance and the concept of cause are 'a priori conditions for the possibility of experience'.

This was Kant's response to Hume's fork. Kant saw the possibility of a form of argumentation which is not logical but 'transcendental' (hence the rather cumbersome phrase 'transcendental deduction of the categories'). Transcendental argument takes as a datum the fact that we have 'experience' of some form or other, then inquires how this is possible. Again, it is a matter of some controversy whether transcendental argument, in Kant's hands, is just a more sophisticated application of logic.

Kant wrote the Critique of Pure Reason to defend metaphysics. However, he accepted the broader implications of Hume's fork that the traditional ambitions of metaphysics were too grandiose. This is in fact what Kant is referring to in the quotation which concludes your essay. Attempts to prove the existence of God (for example) using 'pure reason' can never succeed, Kant thought, and in the second part of the 'Critique' he proceeds to show in detail why. Whereas transcendental argument is tied to experience, in attempting to account for its possibility, pure reason attempts to cut loose from experience altogether and consequently finds itself in irresolvable contradictions.

In the light of this, it is worth asking, how Humean was Kant? Hume worked up quite a sophisticated theory of the human faculty of imagination which on the basis of perceptual data creates a world of 'fictions'. Kant uses his transcendental argument to establish the existence of an external world of objects in space -- but then qualifies this claim drastically admitting that these objects are mere 'phenomena' and human beings can have no knowledge of how things are 'in themselves'. -- You may wonder whether there is that much to choose between Kantian 'phenomena' and Humean 'fictions'.

All the best,

Geoffrey